Talk:Vowel length

English Phonics (/juː/ is a syllable, not a vowel)

I made the table of "Traditional English Phonics" but am concerned with some of the ways it has been changed. First is that /e/ and /eɪ/, and /o/ and /oʊ/ are allophones (at least in GenAm), the realization being based on stress. I'm OK with this change, however, due to the allophonic nature. What does bother me is changing /uˑ/ to /juː/ -- the [j] onglide is neither universal nor a vowel (thus not part of the vowel), and /juː/ certainly is not how I was taught this as a kid. More to the point, /juː/ (by itself) is a syllable (not a vowel), and the [j] is a part of the consonant cluster (even if not shown orthographically).

(Also, the table was not intended to highly silent e specifically, but I see no problem with that, either).

Does anyone else agree that "long u" is /uˑ/ not /juː/? --Jared (talk) 01:01, 12 April 2009 (UTC)

Even more to the point, this is about how it was taught, and (as I remember), it was taught that /u/ was a vowel, it was not taught that /j/ is any part of a vowel. --Jared (talk) 01:08, 12 April 2009 (UTC)

It is not unusual to consider /juː/ a diphthong. Many languages have diphthongs starting or ending in [w] or [j], often spelled as u and i or y, or not spelled explicitly. Also the allophonic nature of /juː/ and /uː/ points to the fact that the /j/ is part of the vowel, c.q. diphthong, not a separate syllable initial. −Woodstone (talk) 08:45, 13 April 2009 (UTC)

Whether /ju:/ is phonologically a vowel, a diphthong, or a sequence of glide + vowel isn't even relevant to the table in question. In pedagogical "phonics", /ju:/ is treated as a vowel, specifically the "long u sound". The "long" versions of all the vowels are held to be the same as the name of the letter that represents them: "long A" is /eɪ/, same as the name of the letter A; "long E" is /i:/, same as the name of the letter E; and "long U" is /ju:/, same as the name of the letter U. As the cub/cube example shows, adding a "silent E" changes the "short U" (/ʌ/) into "long U" (/ju:/, not /u:/). —Angr 18:25, 13 April 2009 (UTC)

But the term "long u" is also used for the vowel in "suit" and "lute," and these aren't completely allophonic in English (though I can't think of a minimal pair, /lju:t/ would not be equated to word "lute," nor /ku:t/ for "cute"). The pedagogy I'm familiar with said a "long vowel" was followed immediately be another vowel, or one consonant and a silent e. Perhaps its taught differently in different places or schools, though -- I have no idea what was more common. --Jared (talk) 23:22, 13 April 2009 (UTC)

Actually, in dialects that allow /j/ after /l/, /lju:t/ is the pronunciation of "lute", as distinct from /lu:t/ "loot". The loss of the /j/ in /ju:/ after different consonants in different dialects is entirely predictable. —Angr 06:07, 14 April 2009 (UTC)

That does, however, if these are pronounced differently, produce a sort of minimal pair.--Jared (talk) 23:05, 12 July 2010 (UTC)

There's "mute" and "moot", "beaut" and "boot". I agree with Angr that traditionally /juː/ has been called "long U", which is treated as a vowel inspite of the argument that it's not really just a vowel at all (though as Woodstone notes, it's possible to consider it as being one; also it's not necessarily a syllable either, it is in "ewe" & "yew" but not in "cute"). If you were taught differently, perhaps you had more enlightened teachers. Yes, you might call /uː/ "long u" but it's primary school phonics were talking not tertiary school linguistics. JIMp _talk·cont 00:19, 20 April 2012 (UTC)

Persian anyone?

Someone should add a section to this article regarding the reconstructed vowel-quantity contrasts of Classical Persian, as well as the confusion over whether modern Iranian Persian has contrastive vowel-quantity. I'll do it myself if there's no one around, but I'd rather it be done by a real Persianist (as opposed to an Arabist such as myself who dabbles in Persian.) Szfski (talk) 23:56, 22 August 2009 (UTC)

Do it, I'll check it. Alefbe (talk) 18:34, 23 August 2009 (UTC)

Traditional definitions are WRONG

Why is there nothing in this article stating that the traditional definitions people are taught in school are just plain WRONG. most of the time the teachers teaching it know nothing about linguistics, that should be in the article as well. Also why don't they teach the linguistic definition in grade school? Because I went through most of my life thinking these were long and short vowels until I got into linguistics and found out my idiot teachers had been lying to me for years. Thess issue should be addressed in the article. —Preceding unsigned comment added by Htahpoahf (talk • contribs) 23:54, 18 February 2010 (UTC)

Well, they're not entirely wrong. The article does mention the differences. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 18:58, 19 February 2010 (UTC)

Now we get into semantics -- this is not a "wrong" definition, just a different usage with a different etymology coming from a different field. --Jared (talk) 23:07, 12 July 2010 (UTC)

(Redacted) It's a non-linguistic terminology used to teach children, and it doesn't even refer to sounds but letters. It doesn't take a linguistic genius to notice that i in bite is not just a longer version of i in bit. — Preceding unsigned comment added by 188.147.35.152 (talk) 16:15, 23 October 2018 (UTC)

It's like getting angry at your elementary school teacher for telling you that plants "produce food with air and sunlight" because you learned 10 years later than in fact what they do is reduce carbon dioxide to ribulose-1,5-bisphosphate by knocking electrons out with photons. — Preceding unsigned comment added by 188.147.5.17 (talk) 17:10, 24 October 2018 (UTC)

I disagree. I think the use of the terms "long" and "short" vowels in describing English is wrong. The vowels that are described as being long and short versions of one another are actually different vowels. The letter "i" in the words "kit" and "kite" represents two different vowels, /I/ and the diphthong /ai/, respectively. — Preceding unsigned comment added by Nealmcgrath (talk • contribs) 01:59, 27 October 2013 (UTC)

It's not that uncommon for languages to be described as having vowel length even when short and long vowels have different qualities. Swedish is a good example of that, as is Hungarian. Swedish vowels matche English ones in several cases, in particular the pair /iː and /ɪ/. So the length does not need to be contrastive in order for it to be recognised as length. CodeCat (talk) 02:33, 27 October 2013 (UTC)

I agree that it is wrong. If you say a short vowel more slowly it does not sound like a long vowel. I have been voicing the long and short vowels a bit and have noticed that I place my mouth in different locations when saying long versus short vowels. There are even more pseudo-vowels in English that are traditionally written with more than one latin alphabet character. One of the ten single-latin alphabet character vowel phonemes, long u, is often paired with a slight voiced consonant 'y' in front of it. Ask yourself if something that could be chaotically written as 'ooniverse' and 'universe' sound identical. I am of the mind that the words 'long' and 'short' were created not by linguists but prior to linguistic analysis of the term, in order to describe the chaotic adoption of different vowels from other languages. As far as long versus short vowels following voiced versus unvoiced consonants, I am of the mind that English is chaotic when it comes to multisyllable words and does not tend to follow standard rules very well. It is difficult to say if the article's main assertions are traditional views of linguists, or of non-linguists describing the assimilation of foreign vowels prior to linguistic analysis, since much of the article's key points in this area are unreferenced.75.175.58.72 (talk) 20:56, 24 June 2014 (UTC)

The long-short terminology is not strictly true for English but it can be a useful educational tool because there are many cases of alternations between related pairs. Compare for example divine and divinity. Of course these are historically long-short pairs, but as far as I know this kind of alternation is still slightly productive in English even today. As for universe, it is generally transcribed with [ju-] in IPA. CodeCat (talk) 21:44, 24 June 2014 (UTC)

The basic problem is the article then goes on and compares with other languages but then acts as if the weasel word definition 'perceived' duration does not produce real problems and intrinsic unclearness. In some languages, the duration of time that a vowel is spoken can be a phoneme that changes the meaning of the word spoken. There might even be a few examples in English where this happens but I can not think of any examples at the moment. This however is a different phenomenon from having two distinct different sounds produced for most vowel symbols, at least in written languages that have vowel symbols. Failure to clarify these distinctions in meaning then garbles the clearness of what is being referred to in those sections that deal with language comparisons71.34.84.153 (talk) 19:18, 1 July 2014 (UTC)

Just to be clear, there is a very clear linguistic definition of "long" and "short" vowels, which is (as you may expect) saying the same vowel for a longer period of time makes it long, for a shorter period of time makes it shorter. The commonly used explanation in US grammar schools that the "a" in "mat" is short and the "a" in "mate" is long is incorrect. I am not familiar with specific exampled from languages other than Japanese, but in Japanese, long and short vowels are exactly that: Longer and shorter forms of the same vowel sound, and they are not allophones, they give different meaning to the words. "Chizu" means "map," "chiizu" (with the "/i/" sound said for a longer period of time, means "cheese." Please leave the explanation that the commonly used terminology in English is incorrect in the text! English does not have long and short vowels. Thank you! — Preceding unsigned comment added by Nealmcgrath (talk • contribs) 19:22, 9 May 2016 (UTC) Nealmcgrath (talk) 15:02, 19 July 2020 (UTC)

December 2017: Wow, I and several others who clearly are trained in linguistics have fixed the incorrect explanation of alleged long and short vowels in English many times, and yet someone (or different people) keep going in and putting back the incorrect information.

The be clear, English DOES NOT HAVE LONG AND SHORT VOWELS. Period. This is linguistic fact. Yes, some dialects may exist in which vowel length gives words different meanings, such as Australian English, but standard written English does not have long and short vowels.

As explained, the middle school explanation that the "i" in "kit" is short and the "i" in "kite" is long is WRONG. Those are two different vowels, /I/ and /ai/.

Please stop changing this. — Preceding unsigned comment added by Nealmcgrath (talk • contribs) 16:19, 19 December 2017 (UTC)

Then maybe cite some sources instead of shouting. — Preceding unsigned comment added by 188.147.35.152 (talk) 16:10, 23 October 2018 (UTC)

"The Earth is round and goes around the sun, not flat and the center of the solar system," shall I cite a reference for that statement as well? I can probably find something that does verify that, but do I have to? Seriously, anyone with even basic training in linguistics will know that English does not have long and short vowels and that what is taught in school is not correct! But if you insist, I will try to find a book that explains what everyone in the field knows already. — Preceding unsigned comment added by Nealmcgrath (talk • contribs) 23:24, 18 July 2020 (UTC)

This polemical stuff about vowel length in English is based on some fundamental misunderstandings about the phonology and phonetics of English, and your apparent rewriting of the corresponding section of the article on RP has made the confusion worse. Please make yourself known as a Wikipedia contributor so that issues may be discussed on your talk page. I will be suggesting ways of correcting the misunderstandings, but in the meantime please accept that you may be wrong. RoachPeter (talk) 11:09, 19 July 2020 (UTC)

My apologies if it comes across as "polemic," however, I have been correcting this section literally for years and others continue to change it back to contain incorrect information, it becomes somewhat frustrating. Just be be clear, the issues with the ongoing reversions are: English does not have long and short vowels in the sense that vowel length does not change the meaning of the word. Vowel length may vary depending on other factors, but it does not affect meaning as it does in some other languages. Also, the traditional explanation that the letter "a" in "latter" is "long" and the letter "a" in "later" is "long" is not correct: They are not long and short versions of the same vowel, they are different vowels. This is not a matter of opinion, it is established linguistic fact. I would have difficulty finding a source to cite for this because every linguist knows this, I learned this within the first semester earning my degree in applied linguistics: There will not be a book, paper etc that explains the practice of calling English vowels "long" and "short" is incorrect because every linguist knows this, it is not the subject of discussion let alone debate (except here, apparently) so no one would have bothered publishing a study to explain it. Like I said, it would be like publishing a paper today refuting the idea the Earth is flat. However, this video explains it neatly: https://www.youtube.com/watch?v=GQa9w__GqLc. Not a scholarly source, but explains it quite clearly for laypeople. — Preceding unsigned comment added by Nealmcgrath (talk • contribs) 14:50, 19 July 2020 (UTC) Nealmcgrath (talk) 15:02, 19 July 2020 (UTC)

As many contributors to this Talk section have pointed made clear, there is a fundamental case of crossed wires here. You are reading the "long"/"short" issue from the viewpoint of what is taught in classroom materials for reading. The rest of us are treating this as a topic in English phonology. You are evidently angry about people who link together /æ/ 'mat' and /eɪ/ 'mate' in reading schemes and call them "short" and "long", but the rest of us are concerned about whether or not it is ok to link /æ/ 'mat' with /ɑ:/ 'mart'. There is a perfectly respectable argument for pairing such vowels as /iː/ and /ɪ/ as long and short partners and ignoring the quality difference for the sake of an easy-to-grasp exposition of the English (RP) vowel phoneme system. You might not buy that argument, nor may I, but it is not in the same league as your flat-earth argument. I have modified the article to keep the two approaches to long and short separate, and hope that now settles the matter. RoachPeter (talk) 15:18, 19 July 2020 (UTC)

Sorry, this will not do. You say "In English, length does not change the meaning of the word and what are often described by grammarians as "long" and "short" vowels in English are actually different vowels, not longer and shorter versions of the same vowel." But it has nothing to do with grammarians, and nobody as far as I know is claiming that paired long and short vowels are "longer and shorter versions of the same vowel".

You then say "Schools in English-speaking countries teach that the letter "a" in "later" (actually the sound /e/) is a "short" vowel and the same letter in "latter" is long (it actually represents the sound /æ/)." This is an unsubstantiated claim, and untrue: I have taught in schools in English-speaking countries and never heard this. What do you mean by "actually the sound /e/"?

A passage about the historical development of English vowels and the Great Vowel Shift in the context of school teaching of reading doesn't seem relevant, and you give no references. Almost everything here is just private speculation. RoachPeter (talk) 19:35, 19 July 2020 (UTC)

Since Nealmcgrath has reinstated the material I have been objecting to, I have requested the Wikipedia disputes procedure to resolve the problem. RoachPeter (talk) 09:14, 20 July 2020 (UTC)

Vowel sounds that don't fit into the "short" / "long" dichotomy

What's the terminology for an "a" pronounced as /ɑ/ (as in "father") or /ə/ (as in "about")? DanBishop (talk) 05:28, 21 December 2011 (UTC)

Well, the difference between short and long is vowel quantity, and the difference between a in father and about is vowel quality. And the a in about is a reduced vowel, in contrast to the a in father, which is a full vowel. Is that what you're looking for? — Eru·tuon 05:43, 21 December 2011 (UTC)

In short, these are altogether different vowels. English (almost all varieties) doesn't have true vowel length distinctions, so stop looking for them. --vuo (talk) 00:18, 22 December 2011 (UTC)

You don't have to stop looking for them, just look in the right place. Some dialects of English do have phonemic vowel length as mentioned in the article. For me the vowels spelt with "a"s in "father" and "about" differ in both length (quantity) and position (quality); however, the "a" in "father" vs the "u" in "cut" is a length distinction. JIMp _talk·cont 02:32, 20 April 2012 (UTC)

In what dialect? In most dialects, the "a" in "father" is /ɑ/ and the "u" in "cut" is /ʌ/—they differ in quality. Curly Turkey (gobble) 04:44, 24 April 2013 (UTC)

Only in Australian English, though; I think that's the point that vuo is making. — Preceding unsigned comment added by 130.102.158.16 (talk) 05:23, 3 July 2012 (UTC)

Dubious

The article says that "In American English vowel length is phonemic before the alveolar flap, in minimal pairs such as ladder/latter and liter/leader." As a native speaker of Pacific Northwest English and a student of linguistics, I see no such distinction. Does anyone have further evidence to support or to contradict the statement? — Preceding unsigned comment added by Jeremy Reeder (talk • contribs) 01:42, 20 September 2012 (UTC)

As a native speaker of New England English, I see no such distinctions, especially in the given examples. However, the closest thing I know of is in the /aɪ/ and /aʊ/ phonemes (rider vs. writer and clouding vs. clouting) in some accents including mine, but these are accompanied by strong changes in vowel quality (rider = [ɹaːɪ̯ɾə(ɹ)] but writer = [ɹʌɪ̯ɾə(ɹ)] and clouding = [kɫæːʊ̯ɾɪŋ] but clouting = [kɫɛʊ̯ɾɪŋ]). --- Wikitiki89 (talk) - 15:03, 14 November 2012 (UTC)

The latter is know as Canadian raising, and is a feature of most Canadian English dialects. Curly Turkey (gobble) 04:46, 24 April 2013 (UTC)

Put redirect to English orthography?

I came to this page doing a search for "long vowel", I was searching for an explanation of how this term is used in when talking about English orthography. While this page does cover this use, it is pretty far down on the page. I feel like this is a separate concept from the concept of synchronic vowel length in languages like Japanese and Korean, and should therefore be addressed by a separate article, or by a section in the English Orthography article, rather than here. — Preceding unsigned comment added by SpiralTurtle (talk • contribs) 05:24, 18 November 2013 (UTC)

English vowel lengthening/shortening

In the section 'Short and long vowels in English: Allophonic vowel length' the situation is stated the wrong way round. It is generally accepted by most phoneticians that fortis consonants shorten a preceding vowel, rather than that lenis consonants lengthen one. That's why we refer to pre-fortis clipping when we observe the shortening of, for example, /aɪ/ in 'bite' relative to the same diphthong in 'buy'. The generative rules displayed in this section, as well as stating the facts the wrong way round, are also quite unnecessary: all they say is that voiced consonants lengthen preceding vowels. I also disagree with the last bit of the concluding sentence " the currently prevalent view tends to emphasise the latter rather than the former". So if nobody objects, I intend to rewrite this section. RoachPeter (talk) 09:36, 29 March 2015 (UTC)

If you've got the sources to back it up, you don't need to ask permission to improve the article. Curly Turkey ¡gobble! 11:37, 29 March 2015 (UTC)

Yes, I do have the sources, and I see what you mean, but if I'm going to alter someone else's work in a major way I think it's worth checking that there's no objection. RoachPeter (talk) 15:06, 29 March 2015 (UTC)

I'm not sure if I object, but I'm curious, how do you tell whether it's pre-lenis lengthening or pre-fortis shortening? Do you measure the duration of a pre-fortis vowel and a pre-lenis vowel in seconds or milliseconds, and compare these durations with the duration of a vowel before a sonorant (which is neither lenis nor fortis), or the vowel in some other environment? — Eru·tuon 18:07, 29 March 2015 (UTC)

@Erutuon: the answer is probably here. Peter238 (talk) 21:35, 2 April 2015 (UTC)

@Peter238: Thanks for the link. Still, it only discusses length before fortis and lenis consonants, and leaves the question of what the "default" length is and where it occurs (before sonorants or syllable-finally, perhaps). — Eru·tuon 00:15, 3 April 2015 (UTC)

@Erutuon: The answer is that the default length is the pre-lenis one, as in bead. Its duration is the same as the vowel in bee and about twice as long as the vowel in beat. Pre-lenis lax consonants, such as /ʊ/ have about the same duration as the pre-fortis tense consonants, such as /uː/. My source is the latest edition of Gimson's Pronunciation of English, section 8.4 Vowel length. You should be able to access it through Google Books. Peter238 (talk) 01:47, 3 April 2015 (UTC)

Another source: Patricia Ashby - Understanding Phonetics (2011), section 7.1.2.2 Pre-fortis clipping. It may be available for free online, I'm not sure. Try Googling the title. Peter238 (talk) 02:32, 3 April 2015 (UTC)

Help with long vowels and glottal stops?

In the introduction, it was stated that "However, some languages with two vowel lengths also have words in which long vowels appear adjacent to other short or long vowels of the same type: Japanese hōō "phoenix", Estonian jäääärne "(on the) edge of the ice", or Ancient Greek ἀάατος [a.áː.a.tos] "inviolable". Some languages that do not ordinarily have phonemic vowel length but permit vowel hiatus may similarly exhibit sequences of identical vowel phonemes that yield phonetically long vowels, such as Georgian გააადვილებ [ɡa.a.ad.vil.eb] "you will facilitate it"." However, this is not true for the Estonian example at least. The Estonian example is pronounced somewhat like [jæːʔæːɾne] (Pronunciation example here). There is a glottal stop in the middle, so it is false to say that "vowels [--] appear adjacent to [--] vowels". I removed the Estonian example. The whole section seems misleading and does not take into account the existence of glottal stops. I would guess that Ancient Greek and Georgian probably utilize glottal stops to distinguish between VV-type syllables too, but since I'm not sure about the pronunciation of Japanese, Ancient Greek and Georgian, how should I go about correcting the article? 193.167.41.1 (talk) 12:52, 19 September 2016 (UTC)

The glottal stop is not phonemic in Estonian or any of those other languages, so it's not misleading. The glottal stop is only a phonetic feature. CodeCat (talk) 15:32, 19 September 2016 (UTC)

Allophonic vowel length

There is a serious problem with this section. It is a fallacy to state that "vowels are lengthened before a voiced consonant", and no citation is given for this claim. It has been established scientifically by Chen (1970) and many others since that the effect here is one of the shortening of a vowel when it is followed by a voiceless (fortis) consonant. As a result the realization rule given is wrong and needs to be removed. I will do this, leaving a short time for anyone to object. Incidentally, the same fallacy is seen in English phonology#allophones of vowels. Another objection to this section is that although it is labelled "Allophonic vowel length", the last part is actually about phonemic vowel length differences. RoachPeter (talk) 08:15, 10 April 2018 (UTC)

Now rewritten RoachPeter (talk) 14:42, 12 April 2018 (UTC)

s

Can someone rewrite the "Short and long vowels in English" section? It was written by the angry idiot above without citing any sources. It repeats multiple times that "the common explanation" is INCORRECT, and WRONG, unnecessarily shouting at the innocent reader. Well, this "what is taught" is not taught to students of linguistics, but to kids in schools, and they don't learn phonetics, they learn how to write - the term "vowel" doesn't even refer to a sound in this scenario, but to a letter. No one has ever claimed that "e" in Pete is just a longer version of "e" in pet. It's like getting angry at your Italian teacher who told you that <c> in cena is "dolce", and then you tried to eat the paper and it wasn't sweet at all. — Preceding unsigned comment added by 188.147.35.152 (talk) 16:22, 23 October 2018 (UTC)

Revert

Nardog may I know what is your goal in reverting my disclaimer about an unsourced section written by an angry man shouting every third word how incorrect something is? (Redacted) — Preceding unsigned comment added by 188.147.35.152 (talk) 17:18, 23 October 2018 (UTC)

The main purpose of my revert was to undo Nealmcgrath's edits, not yours. If you still think the templates you added ({{tone}} and {{original research}}) apply regardless, keep them. Nardog (talk) 17:25, 23 October 2018‎ (UTC)

The edits you undid made the section incorrect again. The shouting man is right: English does not have long and short vowels, it is established linguistic fact, to say it is unsourced is like saying the Earth is round needs to be sourced, anyone with even basic knowledge of linguistics knows this. The explanation that the "e" in "bet" is a short version of the "e" in "beet" is wrong, they are different vowels, not long and short versions of the same vowel. Why people keep changing it to be factually incorrect over and over again is beyond me. Please find yourself a book "intro to linguistics" or something like that and read up on phonemics and pay particular attention to the section on long and short vowels. — Preceding unsigned comment added by 73.199.213.24 (talk) 22:46, 18 July 2020 (UTC)

The vowels in merry and Mary contrast in length alone. So much for basic knowledge. Rua (mew) 12:33, 19 July 2020 (UTC)

IPA vowel length image

Nice image at the top of the article, but what does it mean? Could someone move it to a section where the symbols and jargon in the image are explained? I don't know how to do that. DBlomgren (talk) 17:43, 15 May 2019 (UTC)

RP / contemporary southern British

In contemporary southern British standard speech, or what might be called "realistic RP", the following phonemes are distinguished mainly or exclusively by vowel length: (1.) /ɪ/ and /ɪə̯/ as [ɪ] vs. [ɪː], (2.) /ɛ/ and /ɛə̯/ as [ɛ] vs. [ɛː], (3.) /ə/ and /ɜː/ as [ə] vs. [əː]. This is mentioned under Cockney, which is clearly wrong... You might call it Estuary English, but I'd say it's pretty much the contemporary standard pronunciation, isn't it? 77.191.75.203 (talk) 15:46, 28 February 2021 (UTC)

Directed to section on spelling and teaching reading

There is one notice, that "Short i" is directed here. But there are several similar titles which direct here, like "Long u". It would be cumbersome to make a notice for each of those. Is there some way to take account of all of these - I dare not suggest a generic "long/short vowel letter directs here" and BTW, somehow to direct to the Cyrillic letters, in cases when they exist. TomS TDotO (talk) 13:07, 27 April 2023 (UTC)